15 results about "Spatial fourier transform" patented technology

The invention provides a microwave detection method and device for PE pipe hot-melting weld defects, and the method comprises the following steps: 1) taking a hot-melting weld of a PE pipe and a base material near the weld as detection objects, and carrying out axial and circumferential microwave scanning to obtain a reflection coefficient of each position point; 2) carrying out two-dimensional space Fourier transform on the reflection coefficient of each position point; 3) carrying out phase correction on the converted data in the depth direction; 4) carrying out two-dimensional space inverse Fourier transform on the corrected data to obtain a reflection coefficient of the target; 5, image reconstruction is carried out on the reflection coefficient of the target, and information of the defects in the PE pipe hot-melt welding seam is obtained.According to the scheme, the method has the advantages that the defects can be detected, the defect positions can be determined, and conclusion obtaining does not excessively depend on personnel skills.

The present invention provides a system and method for calibrating the geometric features of three-dimensional interference fringes. The system combines a high-precision linear motorized translation stage with a matrix camera. A control program is used to manipulate the translation stage along the projection direction of the interference fringes, while the camera is controlled to capture interference fringes at different locations to produce a three-dimensional image of the fringes. A spatial Fourier transform, Gaussian fitting, and a maximum position extraction algorithm are used to calculate the spatial frequency of the variations in light and dark fringes, thereby calibrating the fringe spacing of the interference fringe system based on pixel size. Image binarization and an edge detection algorithm are used to extract the fringe boundary contours, calibrate the height and width of the interference fringe area, and ultimately evaluate the geometric features of the three-dimensional interference fringe system. This system and method can provide a more comprehensive calibration of interference fringe systems.

The invention discloses a seismic wave field simulation method and system in an attenuation medium. A new constant Q viscous sound wave equation is constructed to accurately meet the dispersion relation of the Kjartansson constant Q theory, and an amplitude attenuation item and a phase dispersion item are subjected to explicit separation, so that complete decoupling of attenuation and dispersion can be realized while the theoretically accurate constant Q dispersion relation is kept, and the frequency dispersion accuracy is improved. And solving a constant Q viscous acoustic wave equation through a generalized pseudo-spectral numerical simulation method based on space Fourier transform to obtain a seismic wave field with spatio-temporal variation, so that the precision of viscous acoustic wave simulation is remarkably improved.

The invention provides a method for determining a control filter for sound field partitioning and a sound field partitioning control method, and relates to the technical field of acoustic signal processing. A sound field constructed by a complex transmission path can be converted into a limited low-order modal coefficient to represent a main propagation mode of the sound field. The low-order modal coefficient adopted in the method is insensitive to environmental disturbance such as noise, temperature and humidity, and has higher stability and robustness. According to the method, each low-order modal coefficient is introduced into the cost function, so that even if a certain error exists in the determination process of the transmission path, the sound field partitioning effect of the control filter cannot be obviously influenced, and the control filter has robustness of resisting mismatch of the transmission path while ensuring the sound field partitioning effect.

A deep learning framework, termed Fourier Imager Network (FIN) is disclosed that can perform end-to-end phase recovery and image reconstruction from raw holograms of new types of samples, exhibiting success in external generalization. The FIN architecture is based on spatial Fourier transform modules with the deep neural network that process the spatial frequencies of its inputs using learnable filters and a global receptive field. FIN exhibits superior generalization to new types of samples, while also being much faster in its image inference speed, completing the hologram reconstruction task in ˜0.04 s per 1 mm2 of the sample area. Beyond holographic microscopy and quantitative phase imaging applications, FIN and the underlying neural network architecture may open up various new opportunities to design broadly generalizable deep learning models in computational imaging and machine vision fields.

The invention discloses a method for extracting surface wave dispersion of vehicle source signals under the condition of low road flatness, and relates to the technical field of vehicle source signals. Each channel of a power spectrum function calculated from a vehicle source signal is multiplied by an offset distance r to obtain a new function. A spatial Fourier transform of the new function with respect to the offset distance r is performed to obtain a frequency wavenumber spectrum. The frequency wavenumber spectrum F(ω, k') is interpolated and superimposed to obtain the method. Based on the F‑J method, the invention improves the calculation method through the characteristics of equidistant linearly arranged detectors used in vehicle source signal acquisition, and improves the efficiency and accuracy of dispersion curve extraction.

A radar evaluation system for distinguishing objects in the vicinity of a vehicle is disclosed, wherein the radar evaluation system includes a central evaluation unit which is set up as follows: - to receive a first set of range-Doppler power spectra generated as a function of a first set of received signals generated by a first set of receiving antennas of a stereo radar system, - to receive a second set of range-Doppler power spectra generated as a function of a second set of received signals generated by a second set of receiving antennas of the stereo radar system, - to select a pair of frequencies for each power spectrum of the first and second sets of power spectra, wherein the respective power spectrum of the first and second sets of power spectra includes phase information of the selected pair of the respective power spectrum, - Performing a spatial Fourier transformation based on the phase information of the selected pairs. - to input a result of the spatial Fourier transformation as input data into a neural network and to receive an output from the neural network as a response, where the output indicates a distinction between the objects.

The present invention discloses a density stratification flow velocity correction method and system based on the linear multi-layer non-hydrostatic theory, including: obtaining the seawater density distribution and water level change distribution of a target sea area, and calculating the dimensionless density profile of seawater based on the seawater density distribution; stratifying the target sea area longitudinally to obtain a number of seawater layers, and calculating the dimensionless density and dimensionless thickness corresponding to each seawater layer based on the dimensionless density profile of seawater; calculating the flow velocity attenuation coefficient based on the total number of seawater layers, dimensionless density and dimensionless thickness; performing a spatial Fourier transform on the water level change distribution to obtain the wave number domain distribution of the water level change; calculating the wave velocities at different wave numbers based on the wave number domain distribution according to the linear wave theory; correcting the wave velocities based on the flow velocity attenuation coefficient, and outputting the corrected wave velocities.

The application discloses a dynamic equalization and howling suppression intelligent tuning method of a digital audio processor and belongs to the technical field of computers. The method is characterized in that a distributed high-fidelity microphone array is laid out, a wideband sweep excitation signal is output, and sound pressure data are synchronously collected; potential howling modal regions are extracted through spatial Fourier transform and cross-spectral density analysis, real resonance modes are confirmed through modal matching combined with finite element simulation; the critical state of howling is determined by monitoring the growth of modal energy and the spatial deviation, and finally, local equalization filtering is activated according to the howling position. The application realizes accurate positioning and local suppression of howling, maximally retains the original sound quality, and improves the stability and adaptive ability of an audio system in a complex acoustic space.

The invention discloses a method for realizing particle annular control based on a double-ring perfect vortex sound field, which comprises the following steps of: performing phase modulation on sound waves emitted by a planar fan-shaped transducer array by designing a special sound lens to generate coaxially propagated Bessel type composite vortex sound beams with different radial wave numbers; a double-ring perfect vortex sound field is formed through spatial Fourier transform; the radial wave number is adjusted through lens structure parameters to achieve accurate regulation and control of the double-ring perfect vortex radius, the sound pressure and the phase, stable capture of conventional particles between double-ring perfect vortex sound fields and one-way movement of the conventional particles along an annular low-sound-pressure track are achieved, and regulation and control of the movement speed of the particles are achieved. The method breaks through the application limitation that a traditional vortex sound field center capture and a single-ring perfect vortex sound field can only capture negative acoustic coefficient particles, and has the advantages that the ring radius, the capture range and the movement speed are flexible and adjustable, the particle control stability is high, and the system robustness is high; and a new solution is provided for the practical application of the sound control technology in the fields of biomedical engineering and the like.

The invention relates to the field of laser lighting, in particular to a light mixing device and a lighting device.The light mixing device comprises a multicolor light source used for emitting multiple beams of incident light with separated space and wavelength; multiple beams of incident light enter the first light guide column at a preset incident angle, and a light field of a seed spectrum is formed at the emergent end of the first light guide column through internal total reflection; the optical Fourier transform lens is used for performing spatial Fourier transform on the light field containing the seed spectrum; the light incident surface of the second light guide column is arranged near the focal plane of the optical Fourier transform lens, receives the light field subjected to Fourier transform and spreads spectrum distribution through internal total reflection, so that emergent light forms a uniformly distributed spectrum in a two-dimensional space; according to the scheme, through a physical mechanism of seed spectrum generation-Fourier transform-spectrum expansion, the problem that light sources with different light beam qualities completely coincide in space is solved, and finally light beams which can be uniformly mixed in a near field and a far field and have excellent chromaticity uniformity are obtained.

The invention discloses a material sound absorption performance measurement method based on spatial Fourier transform, which comprises the following steps: acquiring acoustic data of limited measurement points of a target material, and outputting spatial sound field data of the target material by utilizing a fitting model based on the acoustic data of the limited measurement points, calculating a sound absorption coefficient and a reflection coefficient of a target material according to the output space sound field data; therefore, the mapping relation between the limited measuring point data and the global complete measuring point data is learned through the neural network model, accurate space Fourier transform can be achieved under the condition that only a small number of limited measuring points are used, acoustic quantities such as the reflection coefficient and the sound absorption coefficient of the material can be calculated through the value output by the model, and the accuracy of acoustic measurement is improved. And the measurement efficiency and the economical efficiency are effectively improved.

The present application relates to a method, device and equipment for decomposing the direction of internal solitary waves based on time-space Fourier transform, and relates to the technical field of marine internal wave signal processing. The method comprises: performing time Fourier transform processing on a parameter time series corresponding to the numerical simulation result of the internal solitary wave obtained, obtaining a target frequency domain signal containing the target positive frequency of the internal solitary wave signal in each fluctuation component, performing two-dimensional space Fourier transform processing on the target frequency domain signal, obtaining target wave number domain information, the target wave number domain information containing the internal wave information in a specific direction, performing inverse Fourier transform on the target wave number domain information, obtaining a target internal solitary wave signal propagating in the target direction. The present application makes full use of Fourier transform, and realizes separation of multi-directional internal solitary waves by filtering the internal solitary wave signal in the frequency domain and the wave number domain, thereby solving the problem that the prior art cannot accurately separate the internal solitary waves.

A semiconductor photonic integrated circuits (PICs) may perform RF signal processing, including beamspace processing or spatial Fourier Transforms in a semiconductor PIC. An RF imaging system including the semiconductor processing PIC may include an antenna array that upconverts RF signals to optical signals using electro-optic modulators, such as lithium niobate modulators. Simultaneous processing (beamforming) of multiple RF signals utilizing the semiconductor processing PIC may be performed.

The present invention specifically relates to a rapid wavenumber domain imaging method for molecular apertures based on a rectangular-boundary MIMO array. The method comprises: receiving an echo signal from a target and preprocessing the echo signal to obtain a corresponding intermediate frequency signal; dividing the intermediate frequency signal based on a subaperture partitioning strategy for the antenna array to determine subaperture data corresponding to each subaperture; performing a two-dimensional spatial Fourier transform on the subaperture data to obtain wavenumber domain data corresponding to the subaperture data; performing subaperture imaging on the wavenumber domain data based on a Fourier transform, and generating a three-dimensional image of the target based on the subaperture imaging. This method can reduce imaging time and significantly improve imaging efficiency.